This shows you the differences between two versions of the page.
Both sides previous revision Previous revision Next revision | Previous revision Next revision Both sides next revision | ||
cluster:208 [2022/05/31 12:25] hmeij07 [gpu testing] |
cluster:208 [2022/06/01 12:51] hmeij07 [gpu testing] |
||
---|---|---|---|
Line 385: | Line 385: | ||
===== gpu testing ===== | ===== gpu testing ===== | ||
- | * test slurm v 21.08.1 | + | * test standalone |
* n33-n37 each: 4 gpus, 16 cores, 16 threads, 32 cpus | * n33-n37 each: 4 gpus, 16 cores, 16 threads, 32 cpus | ||
* submit one at a time, observe | * submit one at a time, observe | ||
Line 405: | Line 405: | ||
* do all 16 jobs log the same wall time? Yes, between 10.10 and 10.70 hours. | * do all 16 jobs log the same wall time? Yes, between 10.10 and 10.70 hours. | ||
- | * ohpc slurm v | + | * ohpc v2.4 slurm v 20.11.8 |
+ | * part=test, n 1, B 1:1:1, cuda_visible=0, | ||
+ | * hit a bug, you must specify cpus-per-gpu **and** mem-per-gpu | ||
+ | * then slurm detects 4 gpus on allocated node and allows 4 jobs on a single allocated gpu | ||
+ | * twisted logic | ||
+ | * so recent openhpc version but old slurm version in software stack | ||
+ | * trying standalone install on openhpc prod cluster | ||
+ | * do all 4 jobs have similar wall time? Yes on n100 varies from 0.6 to 0.7 hours | ||
+ | |||
===== Changes ===== | ===== Changes ===== | ||